PH13608: CALLING THE SYSHADOOP.LOG_ENTRY FUNCTION RETURNS "$BIGSQL_LOG_DIR IS NOT SET!"

APAR status

• Closed as fixed if next.

Error description

• You call the SYSHADOOP.LOG_ENTRY table function, and it returns
  the following message that the BIGSQL_LOG_DIR is not set
  instead of returning the actual error messages from the bigsql
  logs for the specified log identifier.
  For example:
  db2 "select cast(line as char(200)) from
  table(syshadoop.log_entry('BSL-0-30ed5c900'))"
  1
  ---------------------------------------------------------------
  - $BIGSQL_LOG_DIR is not set!
    1 record(s) selected.
  .
  Note that after applying this fix, the steps below in the
  LOCAL FIX section still need to be followed.  Applying this
  APAR fix will address the problem on any new member nodes
  that are added to the cluster in the future, but the existing
  nodes need to be manually cleaned up.
  

Local fix

• 1/ In the Ambari UI, go to  IBM Db2 Big SQL > Configs >
  Advanced bigsql-env  and note the value for the property "Big
  SQL logging directory" (bigsql_log_dir).   By default, this is
  set to /var/ibm/bigsql/logs
  2/ Add the following line to the /etc/profile.d/bigsqlenv.sh
  file on all Big SQL nodes (note: change the specified directory
  to the value per step 1 if it is different than
  /var/ibm/bigsql/logs):
  export BIGSQL_LOG_DIR=/var/ibm/bigsql/logs
  3/ Edit the ~/sqllib/userprofile file under the "bigsql" user's
  home directory, and add the following line before the final
  "fi", on all Big SQL nodes:
  DB2ENVLIST="${DB2ENVLIST} BIGSQL_LOG_DIR"
  4/ Restart the Big SQL service.
  

Problem summary

• Please see the problem description.
  

Problem conclusion

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels